Glowstone future

  • Contributor


    I’m working with Glowstone almost a year. Now I see some good changes in project (Glowkit with git patches, back to old name, new forum…), but what we should do next?

    I really hope in future Glowstone will be used on 50% servers. But how we can do this?

    I think first we should give reason to switch to Glowstone, mean before implementing all features.
    Why? Less popularity = less devs will help with project.

    How we can get it?

    In my opinion, most important thing is performance. Server’s owners need handle as many as possible players. They don’t care about how it works. Just player amout and TPS.

    My simple suggestions are:

    • multithread core
      • separated for 16x16 chunks
      • separated for chunk list, with specified load
      • tick separated to few thread (players, entity, blocks)
    • add async chunk loading (its really important)
    • base player’s function
    • maybe some forge support (wait a bit for it)

    Why that? It’s base. Change this after implementation will be very, very hard.

    Feel free to talk, and share it with friends 😉

  • How about features like Spoutcraft or the VoxelModPack used to have? I think there’s a hole to fill since they died.

    The idea of these projects was to extend the possibilities to communicate between client and server sothat things like highly customizable GUIs, custom items and mobs with server-side textures and models etc. become possible. I think it’s way easier to maintain something like that nowadays and with a 100% original server instead of the Mojang server.

  • Contributor

    I thought and we talked on IRC about that few times. First we have to rewrite base system, then create a simple mod for forge or vanillia client.

    Main problem is performace. See how Forge runs.

    Associated Sponge thread: sponge new era

  • @kamcio96 Who needs Forge? LiteLoader masterrace 😄

  • Admin

    Liteloader is hugely limited and wouldn’t be suitable for something like this (their docs even state that forge is better for large projects).

    It’s an interesting idea, but we’d be more likely to just add an easier way for plugins to do this.

  • Contributor

    @gdude2002 Sure, special api, some tutorials how port the mod, maybe class transformer for small plugins, we will do it 😉

Log in to reply